This open, prospective study investigated the efficacy of lithium augmentation in elderly patients who had failed to respond to a 6-week course of antidepressant medication. Twenty-one patients (age range, 64 to 88 years) with DSM-III-R unipolar, nonpsychotic major depression refractory to a 6-week course of nortriptyline, fluoxetine, or phenelzine underwent a minimum of 2 weeks of lithium augmentation. Five patients had a complete response, 3 had a partial response, and 13 did not respond. Half of the patients developed dose-limiting side effects, most commonly neuromuscular or neurologic. Patients developing side effects were significantly older (p = 0.05), and there was a tendency for side effects to occur more frequently in patients treated with fluoxetine. The rate of response to lithium augmentation in this elderly population was lower than that reported for most open and controlled studies in younger patients. Possible reasons for this are discussed.